Upload
others
View
9
Download
0
Embed Size (px)
Citation preview
Fano resonance based optical modulator reaching 85% modulation depthWenyu Zhao, Huan Jiang, Bingyi Liu, Yongyuan Jiang, Chengchun Tang, and Junjie Li Citation: Applied Physics Letters 107, 171109 (2015); doi: 10.1063/1.4935031 View online: http://dx.doi.org/10.1063/1.4935031 View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/107/17?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Enhancement of light absorption using high-k dielectric in localized surface plasmon resonance for silicon-basedthin film solar cells J. Appl. Phys. 109, 093516 (2011); 10.1063/1.3587165 Fully complementary metal-oxide-semiconductor compatible nanoplasmonic slot waveguides for siliconelectronic photonic integrated circuits Appl. Phys. Lett. 98, 021107 (2011); 10.1063/1.3537964 Far-field observation of the radial profile of visible whispering-gallery modes in a single microdisk based on Si-nanocrystal/ Si O 2 superlattices J. Appl. Phys. 106, 123102 (2009); 10.1063/1.3273360 Resonance control of membrane reflectors with effective index engineering Appl. Phys. Lett. 95, 023110 (2009); 10.1063/1.3182801 Tunable optical filters with nanostructured suspended gratings J. Vac. Sci. Technol. B 27, 1035 (2009); 10.1116/1.3114465
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
159.226.35.163 On: Mon, 07 Dec 2015 01:24:46
Fano resonance based optical modulator reaching 85% modulation depth
Wenyu Zhao,1 Huan Jiang,1 Bingyi Liu,1 Yongyuan Jiang,1,2,a) Chengchun Tang,3
and Junjie Li31Department of Physics, Harbin Institute of Technology, Harbin 150001, China2Key Laboratory of Micro-Optics and Photonic Technology of Heilongjiang Province, Harbin 150001, China3Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academyof Sciences, P.O. Box 603, Beijing 100190, China
(Received 7 September 2015; accepted 21 October 2015; published online 29 October 2015)
In this paper, we demonstrate the combination of nematic liquid crystal with a binary silicon nanohole
array to realize a high performance Fano resonance based optical modulator. The simulations using a
finite difference time domain method reveal that the sharp Fano profile in the binary array originates
from the interaction of the in-phased and anti-phased lattice collective resonance hybridized through
lattice coupling effects. Experimental results agree very well with the simulations and demonstrate the
strong dependence of the Q factor and spectral contrast of the resonance on the radius difference of the
two nanohole arrays. Infiltrated with nematic liquid crystal, E7, the Fano profile can be dynamically
and continuously tuned by an applied voltage, and an unprecedented modulation depth up to 85% is
achieved. VC 2015 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4935031]
Active plasmonics enabling real-time control over nano-
devices’ properties have attracted a tremendous amount of
attention in the past few years.1–4 Great progress has been
made towards a wealth of promising applications such as
active catalysis,5 optical nanoantennas,6 solar cells,7 and bio-
logical sensing and imaging.8 Among the various active tun-
ing media, such as stimuli-responsive polymers,9
photochromic molecules10 and inorganic phase-transition
materials,11 liquid crystal (LC) is an excellent candidate
because of its relatively large birefringence on refractive
index and versatile driven methods to stimulate molecule
realignment.1,2 Many research works have been conducted to
incorporate the LC into metal nanostructures to manipulate
the transmission or the reflection of the plasmonic structures,
which have potentials for color filters, nano-switches, and
nano-modulators.12–19 However, due to the low Q factor and
poor spectral contrast of the resonances in metal nanostruc-
tures, the modulation depths underpinning most nano-
devices are limited to rather small values posing a great
obstacle for practical applications. Improving the perform-
ance of modulation depth remains a major challenge and
more research efforts are desirable.
Recently, Fano resonances in nanostructures resulting
from the interaction of “dark states” and “bright states” have
been extensively studied.20,21 Due to the high Q factor and
spectral contrast, Fano resonances have been employed in
many applications where sharp resonances are essential for
the performances such as biosensors,22 filters,23 surface-
enhanced Raman scattering (SERS),24 and surface plasmon
lasers.25 More recently, we demonstrated a binary silicon
particle array and studied the multi-trapped modes and corre-
sponding Fano resonances resulting from the lattice coupling
effects in the array.26 Because of the low loss nature of
dielectric,27–29 the Q factor and spectral contrast of such
Fano resonances can be strongly improved simultaneously
resulting in an ultra-high figure of merit that is orders higher
than its conventional counterparts made of metal.
In this paper, we take advantage of the easy tuning mecha-
nism of nematic LC and the high Q factor and spectral contrast
of the Fano resonance in a binary nanohole array perforated in
a silicon film to realize a high performance optical modulator.
Simulations using finite difference time domain method dem-
onstrate that the origin of the sharp asymmetric profile sup-
ported by the binary nanohole array is related to the excitations
of the anti-phased lattice collective resonance (ALCR) and the
in-phased lattice collective resonance (ILCR) hybridized
through lattice coupling effects. Experimental results and sim-
ulations are in very good agreement and show the strong de-
pendence of the Q factor and spectral contrast on the radius
difference between the two nanohole arrays. We then incorpo-
rate commercial E7 LC into the binary nanohole array and
study the performance of such structure as optical modulator.
The Fano profile dynamically shifts towards the short wave-
length as the applied voltage gradually increases from 0 to
26.7 kV/cm, and modulation depth up to 85% can be realized
for the operating wavelength of 713 nm.
Fig. 1(a) shows the artistic impression of the binary nano-
hole array composed of two nanohole arrays with different
radii of R1 and R2 perforated in a silicon film (H ¼ 30 nm).
The periods of the two arrays are both P (400 nm) and the
centers of the nanoholes have a displacement of half period.
In order to fabricate the binary nanohole array, a layer of
amorphous silicon film is deposited using plasma enhanced
chemical vapor deposition (PECVD) method on a glass sub-
strate pre-deposited an indium tin oxide (ITO) layer of 20 nm
as electrode. The binary nanohole arrays with footprints of
100 lm� 100 lm are then fabricated using standard electron
beam lithography followed by ion beam etching process
(Fig. 1(c)). To measure the transmission of the samples, white
light from a halogen tungsten lamp is focused on a spot size
of nearly 50 lm in diameter by a 5� objective, and then on
the other side collected by another 20� objective and trans-
mitted to an optical fiber spectrometer (USB2000þ, Ocean
a)Author to whom correspondence should be addressed. Electronic mail:
0003-6951/2015/107(17)/171109/5/$30.00 VC 2015 AIP Publishing LLC107, 171109-1
APPLIED PHYSICS LETTERS 107, 171109 (2015)
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
159.226.35.163 On: Mon, 07 Dec 2015 01:24:46
Optics). The simulations are conducted using commercial
finite difference time domain method software, FDTD
Solutions from Lumerical, Inc. The permittivity of glass, sili-
con, and ITO are extracted from the experimental data.30,31
A sharp Fano resonance with a high spectral contrast
(Fig. 1(b)) appears in transmission around 650 nm, and the
Fano profile evolves from a dip of 4% transmission to a peak
of 65% transmission within 10 nm. Measured transmission
(Fig. 1(d)) agrees very well with the simulation results except
a slight deviation of resonance wavelength mainly due to the
deviations of geometric parameters caused during fabrication
process. The distinct Fano profile can be explained using the
hybridization theory;26 each nanohole array supports corre-
sponding lattice collective resonance which is the lattice col-
lective excitations of the cavity modes supported by the
constituent nanoholes in the array.32 When two nanohole
arrays with different radii are combined to form a binary one,
the two supported lattice collective resonance modes with dif-
ferent resonant energies will hybridize to generate the ILCR
mode and ALCR mode through lattice coupling effects. In the
ILCR mode, two nanohole arrays will always oscillate
towards the same direction either in-phase or anti-phase with
the incident light; thus, the reradiated fields of the two nano-
hole arrays interfere constructively in the far field resulting in
an enhanced radiation to the free space. The ILCR mode is
termed as “bright mode” because of its enhanced scattering
and effective coupling to the incident light. In the ALCR
mode, two arrays will oscillate towards the opposite direc-
tions, leading to the destructive reradiated fields in the far field
and strong elimination of scattering loss. This resonant mode
can be considered the “dark mode” due to its long lifetime
and weak coupling to the free space radiation. The interaction
of the ALCR mode and ILCR mode gives rise to the sharp
asymmetric Fano profile in transmission.
To interpret the ALCR mode and ILCR mode more
clearly, we plot the phase map of the electric field of the bi-
nary array for a unit cell at the plane across the centers of
each nanohole (Z¼ 15 nm). Figs. 2(a) and 2(b) correspond to
the phase map at 660 nm and 740 nm (R1 ¼ 50 nm and
R2 ¼ 110 nm), respectively. The excited cavity modes in the
FIG. 1. (a) Artistic impression of binary silicon nanohole array with different
radii of R1 and R2. (b) Calculated transmission using FDTD Solutions. (c)
Scanning electron microscope (SEM) image and (d) measured transmission of
the fabricated binary nanohole array for R1 ¼ 50 nm and R2 ¼ 110 nm.
FIG. 2. Phase maps of the electric field
of the binary nanohole array for a unit
cell at the plane across the centers of
each nanohole (Z¼ 15 nm) for (a)
ALCR mode and (b) ILCR mode.
Transmission of the binary nanohole
array for different R1, while R2 is fixed
at 110 nm extracted from (c) simula-
tions and (d) experiments. The bottom
row shows the corresponding SEMs of
the binary nanohole array.
171109-2 Zhao et al. Appl. Phys. Lett. 107, 171109 (2015)
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
159.226.35.163 On: Mon, 07 Dec 2015 01:24:46
nanoholes can be regarded as electric dipoles collectively
oscillating with the incident light. The coupling effects
between neighboring nanoholes are relatively weak for the
field mainly concentrating inside each nanohole. When the
incident wavelength is 660 nm (Fig. 2(a)), the ALCR mode
is excited, and the nanoholes in the two arrays oscillate with
a phase difference of p. Because of the same strength of the
resonant amplitude, the reradiated fields of two arrays
strongly cancel each other out in the far field, and the cou-
pling between the structure and free space light is largely
minimized resulting in a Fano dip in transmission. At
740 nm (Fig. 2(b)), the ILCR mode dominates the radiation
with nanoholes in two arrays oscillating in the same phase;
thus, their reradiated fields interfere constructively and
strongly couple with the incident light.
The Fano profiles are closely related to the radius differ-
ence of the binary nanohole array. Figs. 2(c) and 2(d) show
the evolution of transmission spectra as R1 increasing from
40 nm to 110 nm, while R2 is fixed at 110 nm. As R1 gets
closer with R2, the Fano profile gradually diminishes, while
the resonance becomes much sharper with decreased line-
width and spectral contrast. Particularly, for the case of
R1 ¼ R2, the ALCR mode cannot be directly excited by nor-
mal incident light, and the Fano resonance disappears. To
more clearly observe the evolution of the resonance, the
spectral contrast defined as ðIpeak � IdipÞ=ðIpeak þ IdipÞ and
the Q factor is employed to measure the lineshape. The Q
factor defined by Q ¼ xa=2Wa can be calculated by fitting
the transmission spectra to the analytical model33
T ¼
x2 � x2a
2Waxaþ q
� �2
þ b
x2 � x2a
2Waxaþ q
� �2
þ 1
� a2
x2 � x2s
2Wsxsþ q
� �2
þ 1
; (1)
where x is the angular frequency of the incident field, xa
and xs are the resonant frequencies of Fano resonance and
superimposed Lorentz resonance, respectively. Wa and Ws
are the approximations of the spectral widths of both
resonances in frequency units. q and a are the asymmetry pa-
rameter and maximum amplitude of the resonance, respec-
tively. b is the modulation damping parameter originating
from intrinsic losses.33
Fig. 3 shows the evolution of Q factor and spectral con-
trast as R1 gets closer with R2, and the Q factor and spectral
contrast follow different evolution tendencies. As R1 increases,
the Q factor gradually increases (dark line in Fig. 3), indicating
the increasing energy density of the electric field stored in the
resonances of the system. However, the spectral contrast (red
line in Fig. 3) reaches its highest value at small R1 and
decreases as R1 gets larger. This is due to the fact that, in the
binary nanohole array with small radius difference, the energy
density of the electric field is much stronger, and a small loss
introduced by silicon and ITO can result in a large dissipation
of energy, which weakens the Fano profile and reduces the
spectral contrast. In other words, lower R1 means lower energy
density of the electric field stored in the resonance of the bi-
nary nanohole array as well as lower dissipation and higher
spectral contrast. The Q factors and spectral contrasts from
experimental results agree well with the calculations, and the
slight deviation is mainly due to the broadening of linewidth of
the resonances and stray light caused by fabrication defects
and imperfections. The ability of such structure to achieve
high Q factor while still maintaining relatively high spectral
contrast is especially beneficial to improve the performance of
the Fano resonance based devices.
In order to realize Fano resonance based optical modula-
tor, commercial nematic LC, E7, is chosen as the active
media due to its large birefringence (�0.2) and various
driven methods such as electrical field, acoustic wave, heat,
and light. For simplicity, here, we adopt the electric field to
drive the LC molecule realignment. The underlying tuning
mechanism is shown in Fig. 4(a), the binary nanohole array
and ITO glass form a LC cell. The LC molecules exhibit uni-
axial optical symmetry with two principal refractive indices
no and ne. The ordinary refractive index no is for the light
with electric field polarization perpendicular to the director
and the extraordinary refractive index ne is for the light with
electric field polarization parallel to the director. When no
voltage is applied, the rod-like LC molecules reside ran-
domly between the two ITO layers and form a domain struc-
ture, where LC molecules preferably align in the same
direction within one domain and vary randomly from domain
to domain. Macroscopically (in the scale much larger than
the domain size), the LC molecules can be characterized as
an effective media with average refractive index of na.
However, when a bias voltage is applied between the ITO
layer, electric dipoles are formed by one end of a molecule
having a net positive charge, while the other end has a net
negative charge. As a result, the dipole molecules tend to ori-
ent themselves along the direction of the field. Consequently,
the surrounding dielectric environment seen by the binary
nanohole array is actively controlled by the applied voltage.
The change in dielectric environment usually causes a shift
in resonance wavelength, thus, for a certain wavelength,
huge variation of transmission under different applied vol-
tages can be envisioned. In the simulations, the rod-like LC
molecules are modeled as an effective material with refrac-
tive index of n calculated with the help of the macroscopic
order parameter q1
n ¼ na �1
3Dnq; (2)
FIG. 3. Spectral contrast and Q factor evolution as R1 gets closer with R2.
171109-3 Zhao et al. Appl. Phys. Lett. 107, 171109 (2015)
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
159.226.35.163 On: Mon, 07 Dec 2015 01:24:46
where na ¼ ð2no þ neÞ=3 is the average refractive index of
the unaligned LC and Dn ¼ ne � no is the LC birefringence
(ne ¼ 1:734 and no ¼ 1:525 for E7). By applying voltage,
the macroscopic order parameter changes from 0 (unaligned
sample) up to 1 (fully aligned state) and the effective refrac-
tive index varies from na to no, respectively.
Fig. 4(b) shows the simulated transmission spectra for
LC molecules under different macroscopic order parameter q(R1 ¼ 50 nm and R2 ¼ 110 nm). One can notice that the
incorporation of LC results in a sharp Fano profile with nar-
rower width, higher spectral contrast, and resonance wave-
length having a red shift due to the refractive index matching
of the dielectric media on both sides. As the LC molecules
tend to realign along the direction of the applied electrical
field, the effective refractive index seen by the binary nano-
hole array gradually decreases, resulting in a blue shift of the
profile. This is because the light field in the binary nanohole
array is dominated by its X and Y components, as the orien-
tation direction of the LC molecules rearrange from random
state to uniform state along the applied electric field; the
effective refractive index experiences a variation from na to
no. Fig. 4(c) shows the wavelength of the Fano peak for LC
molecules under different macroscopic order parameter q.
10 nm shift of resonance wavelength can be achieved with qchanging from 0 to 1, which is wider than the linewidth of
the Fano resonance.
To experimentally verify the numerical results, the LC
cell is prepared by forming the binary nanohole array with
ITO glass using a 31 lm spacer, and the infiltration process
is done under 80 �C in vacuum environment through capil-
lary action. After infiltration, the sample is slowly cooled
down to room temperature with air pressure recovered to
normal in order to push the LC molecules into the nanoholes.
Figs. 5(a) and 5(b) show the measured transmission spectra
of the binary nanohole array under different applied voltages.
The evolution of the spectra is in good agreement with our
theoretical analysis. The transmission dip blueshifts gradu-
ally from 722.5 nm to 713 nm as the applied voltage
increases, and 26.7 kV/cm voltage is enough to shift the reso-
nance from Fano peak to Fano dip. The Q factor keeps
almost unchanged with the variation of applied voltage (Fig.
5(c)). The maximum modulation depth of 85% is achieved at
713 nm as shown in Fig. 5(d). The transmission monoto-
nously decreases with the increased voltage, and the most
rapid change of transmission is for the applied voltage
around 10 kV/cm. At higher voltage, the variation saturates
FIG. 4. (a) Schematics of active con-
trol of the Fano resonance based opti-
cal modulator. E7 LC molecules
rearrange from random state to uni-
form state under an applied voltage.
(b) Transmission and (c) Fano peaks
for different macroscopic order param-
eter q (R1 ¼ 50 nm and R2 ¼ 110 nm).
FIG. 5. (a) Transmission of the binary nanohole array and (b) its partial
enlargement. (c) Q factors of the Fano resonance under different applied vol-
tages. (d) Transmission at the operating wavelength (713 nm) under different
applied voltages.
171109-4 Zhao et al. Appl. Phys. Lett. 107, 171109 (2015)
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
159.226.35.163 On: Mon, 07 Dec 2015 01:24:46
because the nematic LC molecules are almost completely
aligned along the electric field.
In summary, the Fano resonance within a binary nano-
hole array composed of two nanohole arrays with different
radii are studied both numerically and experimentally.
Simulations using finite difference time domain method
demonstrate that the origin of such sharp resonance is from
the interaction of the hybridized ALCR mode and ILCR
mode. Measured transmission spectra agree very well with
the simulation results and show a different evolution tend-
ency of Q factor and spectral contrast against the radius dif-
ference of the two nanohole arrays. Taking advantage of the
easy tuning mechanism of E7 and the high Q factor and
spectral contrast of the resonance, such binary nanohole
array can act as a high performance optical modulator. The
transmission of the operating wavelength monotonously
decreases with the increasing applied voltage; an unprece-
dented modulation depth of 85% is achieved.
This work was supported by the National Natural
Science Foundation of China under Grant Nos. 50836002
and 51176041. The authors would like to thank the
assistance from the Laboratory of Microfabrication, Institute
of Physics, Chinese Academy of Sciences on the sample
fabrication.
1G. Y. Si, Y. H. Zhao, E. S. Leong, and Y. J. Liu, Materials 7(2), 1296
(2014).2Y. H. Sun, L. Jiang, L. B. Zhong, Y. Y. Jiang, and X. D. Chen, Nano Res.
8(2), 406 (2015).3O. Hess, J. B. Pendry, S. A. Maier, R. F. Oulton, J. M. Hamm, and K. L.
Tsakmakidis, Nat. Mater. 11(7), 573 (2012).4A. D. Boardman, V. V. Grimalsky, Y. S. Kivshar, S. V. Koshevaya, M.
Lapine, N. M. Litchinitser, V. N. Malnev, M. Noginov, Y. G. Rapoport,
and V. M. Shalaev, Laser Photonics Rev. 5(2), 287 (2011).5H. Jing, Q. F. Zhang, N. Large, C. M. Yu, D. A. Blom, P. Nordlander, and
H. Wang, Nano Lett. 14(6), 3674 (2014).6N. Strohfeldt, A. Tittl, M. Schaferling, F. Neubrech, U. Kreibig, R.
Griessen, and H. Giessen, Nano Lett. 14(3), 1140 (2014).7H. A. Atwater and A. Polman, Nat. Mater. 9(3), 205 (2010).8L. Li, U. Steiner, and S. Mahajan, Nano Lett. 14(2), 495 (2014).9A. Baba, K. Tada, R. Janmanee, S. Sriwichai, K. Shinbo, K. Kato, F.
Kaneko, and S. Phanichphant, Adv. Funct. Mater. 22(20), 4383 (2012).
10G. A. Wurtz, P. R. Evans, W. Hendren, R. Atkinson, W. Dickson, R. J.
Pollard, A. V. Zayats, W. Harrison, and C. Bower, Nano Lett. 7(5), 1297
(2007).11Z. L. Samson, K. F. MacDonald, F. D. Angelis, B. Gholipour, K. Knight,
C. C. Huang, E. D. Fabrizio, D. W. Hewak, and N. I. Zheludev, Appl.
Phys. Lett. 96(14), 143105 (2010).12Y. J. Liu, G. Y. Si, E. S. Leong, N. Xiang, A. J. Danner, and J. H. Teng,
Adv. Mater. 24(23), OP131 (2012).13L. Z. Yang, C. Y. Pei, A. Shen, C. Y. Zhao, Y. Li, T. G. Dai, H. Yu, Y.
B. Li, X. Q. Jiang, and J. Y. Yang, Appl. Phys. Lett. 104(21), 211104
(2014).14F. Zhang, X. Y. Hu, C. Y. Wu, H. Yang, and Q. H. Gong, Appl. Phys.
Lett. 105(18), 181114 (2014).15P. R. Evans, G. A. Wurtz, W. R. Hendren, R. Atkinson, W. Dickson, A. V.
Zayats, and R. J. Pollard, Appl. Phys. Lett. 91(4), 043101 (2007).16A. Minovich, J. Farnell, D. N. Neshev, I. McKerracher, F. Karouta, J.
Tian, and Y. S. Kivshar, Appl. Phys. Lett. 100(12), 121113 (2012).17O. Buchnev, N. Podoliak, M. Kaczmarek, N. I. Zheludev, and V. A.
Fedotov, Adv. Opt. Mater. 3(5), 674 (2015).18O. Buchnev, J. Y. Ou, M. Kaczmarek, N. I. Zheludev, and V. A. Fedotov,
Opt. Express 21(2), 1633 (2013).19M. Decker, C. Kremers, A. Minovich, I. Staude, A. E. Miroshnichenko, D.
Chigrin, and Y. S. Kivshar, Opt. Express 21(7), 8879 (2013).20B. Luk’yanchuk, N. I. Zheludev, S. A. Maier, N. J. Halas, P. Nordlander,
H. Giessen, and C. T. Chong, Nat. Mater. 9(9), 707 (2010).21A. E. Miroshnichenko, S. Flach, and Y. S. Kivshar, Rev. Mod. Phys.
82(3), 2257 (2010).22B. B. Zeng, Y. K. Gao, and F. J. Bartoli, Appl. Phys. Lett. 105(16),
161106 (2014).23W. Y. Zhao, X. D. Leng, and Y. Y. Jiang, Opt. Express 23(5), 6858
(2015).24J. Ye, F. F. Wen, H. Sobhani, J. B. Lassiter, P. V. Dorpe, P. Nordlander,
and N. J. Halas, Nano Lett. 12(3), 1660 (2012).25Y. Y. Huo, T. Q. Jia, Y. Zhang, H. Zhao, S. A. Zhang, D. H. Feng, and Z.
R. Sun, Appl. Phys. Lett. 104(11), 113104 (2014).26W. Y. Zhao, D. Q. Ju, Y. Y. Jiang, and Q. W. Zhan, Opt. Express 22(25),
31277 (2014).27B. Slovick, Z. G. Yu, M. Berding, and S. Krishnamurthy, Phys. Rev. B
88(16), 165116 (2013).28J. Cheng, D. Ansari-Oghol-Beig, and H. Mosallaei, Opt. Lett. 39(21),
6285 (2014).29A. Garc�ıa-Etxarri, R. G�omez-Medina, L. S. Froufe-P�erez, C. L�opez, L.
Chantada, F. Scheffold, J. Aizpurua, M. Nieto-Vesperinas, and J. J. S�aenz,
Opt. Express 19(6), 4815 (2011).30E. D. Palik, Handbook of Optical Constants of Solids (Academic Press,
1998).31R. A. Synowicki, Thin Solid Films 313–314, 394 (1998).32A. B. Evlyukhin, C. Reinhardt, A. Seidel, B. S. Luk’yanchuk, and B. N.
Chichkov, Phys. Rev. B 82(4), 045404 (2010).33B. Gallinet and O. J. Martin, ACS Nano 5(11), 8999 (2011).
171109-5 Zhao et al. Appl. Phys. Lett. 107, 171109 (2015)
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
159.226.35.163 On: Mon, 07 Dec 2015 01:24:46